With the increasing popularity of interactive electronic networks, such as the Internet and the World Wide Web, and the advent of digital video technologies, the services of providing digital streaming content via the Internet are drawing more attention in the market place. It is well known that the delivery bandwidth available for transmitting digital video stream or the like to a client home computer is, in general, limited by the network- or Internet-connection speed. As a result thereof, the qualities of the video stream for presentation on a computer monitor or on a large screen display, such as the video frame size and image quality thereof, are spared. Thus, the methods and system needed for compressing digital video data and for decoding and image enhancement is of paramount importance for distributing digital video through the Internet.
The limited data delivery bandwidth available for transmitting digital streaming video and the volume of information represented by such digital images have given rise to the development of various compression “standards”, such as JEPG, H.263, MEPG-2, etc. One basic element of some of these compression methods or standards (e.g., H.263) is to calculate or “predict” the motion of an “image block” in a video stream or image sequence: A video stream is organized hierarchically into groups of image-frame sequence, with each frame including a number of “macro-blocks” for defining the respective portion of the video image therein; An independent frame or “I-frame” is encoded independently of all other frames; Dependent frames are the predictive motion-compensated image frames within a image frame sequence, and are therefore encoded in a manner of depending upon an I-frame and/or on other preceding dependent frames within the same sequence. Since a motion video stream comprises sequences of images that vary incrementally within each sequence, substantial compression can be realized by encoding a number of frames as motion-compensated or dependent frames. The decoding or the reverse process of encoding or compressing the digital images is performed at the client side such that the original image stream can be presented on a display.
Image enhancement generally refers to performing certain operations on video image data to improve image quality of a video stream or a single image frame. Such an operation may include, for examples, contrast manipulation, noise reduction, edge sharpening and/or crisping, filtering, improving digital resolution, magnification and interpolation and the like. It is well known that operations such as transmission over an information network will cause degrading of image signal as a result of noise fluctuation. Prior art image enhancement method, such as the so-called “coring” method or the like, focuses on utilizing filtering (high-pass/low-pass) technique for separating and removing noise from image signal, assuming that noise fluctuations are of relatively small amplitude in comparison to the information components of an image signal, and are concentrated in the high-pass component of the filtered signals.
Although the prior art streaming video compression and enhancement methods outlined above make it feasible for transmitting digital video stream to a client computer over information network having limited delivery bandwidth, such as from a web site to a home computer provided with “dial-up” Internet connection, such compression and enhancement are achieved at the expense of video image signal loss. The prior art video image enhancement method described above only provides means for reducing fluctuating noise signals, and does not improve the quality of the compressed video stream signals at the client side. As a result thereof, a video clip transmitted from a web site using the existing compression and enhancement technologies can only be presented in form of small-image-frame video for display on the computer monitor and be viewed by a user sitting very close to the monitor.
It is known in the art, with respect to the general tendency of information technologies, that the processing speed of personal computing devices such as the PCs will keeps improving, while it would be very expensive or virtually impossible to substantially improve the Internet connection bandwidth of an average household beyond that of the so-called DSL or cable connection. Consequently, there is a need in the art for providing a method of enhancing image signals of a digital video stream by taking advantage of the computing power of a client personal computing device such as the PC.